She had been in anxiety treatment for three years. CBT, two different SSRIs, mindfulness-based approaches. She had worked hard and the treatment had helped, partially. But something wasn't resolving. The panic attacks continued arriving without clear triggers. Her body felt unstable in a way she couldn't account for through psychological work alone.
She was 46. Nobody had asked about perimenopause.
This presentation is not unusual. Research suggests approximately 30% of women experiencing perimenopausal symptoms are initially misdiagnosed with anxiety disorders. The confusion is clinically understandable - but its consequences matter. Years of treatment oriented toward the wrong primary driver produces partial relief at best and, over time, a particular kind of demoralisation: the belief that the problem is simply more intractable than it should be.
Why the confusion is understandable
Hot flashes and panic attacks share almost identical surface presentations. Both can involve rapid heartbeat, sweating or flushing, difficulty breathing, chest tightness, a sense of losing control, and a feeling of impending doom. For a woman experiencing these symptoms - and for a clinician assessing them without specific perimenopause training - distinguishing between a vasomotor symptom and a panic attack is genuinely difficult.
The confusion is compounded by the fact that anxiety symptoms in perimenopausal women are significantly under-researched compared to depressive symptoms. Most research into perimenopausal mood has focused on depression. Anxiety is common during this transition, well-documented in clinical experience, but the evidence base for distinguishing perimenopausal anxiety from primary anxiety disorder is thinner than it should be.
The critical difference: mechanism determines treatment
Despite the overlapping presentation, the mechanisms are distinct - and mechanism determines what treatment is appropriate.
Panic attacks are typically triggered by thoughts, situations, or perceived threat. They are accompanied by catastrophic cognitive content - thoughts about dying, losing control, or going mad. They respond to cognitive and psychological intervention: thought challenging, behavioural experiments, exposure, grounding techniques. Anxiety-specific medications - SSRIs, SNRIs, benzodiazepines - provide relief.
Vasomotor symptoms come from hormonal fluctuation affecting the hypothalamus, the brain's temperature regulation system. They arrive regardless of thought content. A woman can be in a calm, familiar environment with no psychological stressor present and experience a wave of heat, racing heart, and chest tightness. These symptoms do not respond reliably to cognitive techniques, because the trigger is not cognitive.
Treatment for anxiety that doesn't address the hormonal biology providing the substrate for symptoms will provide incomplete relief. Which is exactly what many women describe: the anxiety work is helpful, but something keeps returning that the psychological treatment alone cannot fully reach.
When perimenopause provokes genuine anxiety
The picture is complicated further by the fact that perimenopause can also provoke genuine anxiety - not misdiagnosed anxiety, but anxiety that is neurobiologically driven by the transition itself.
Erratic estrogen fluctuations directly affect the amygdala and the limbic system. Declining estrogen reduces the brain's capacity to modulate the fear response. The result is a lowered threshold for anxiety activation - the nervous system becomes more reactive, less easily soothed, more prone to threat detection. This is anxiety that is both real and neurobiologically rooted in the transition.
In clinical practice, both are often true simultaneously. There may be an underlying anxiety vulnerability that predates perimenopause. There is also a neurobiological amplifier arriving with the transition. Treating only the psychological component, or only the hormonal component, leaves the other layer unaddressed.
Treatment for anxiety that doesn't address the hormonal biology providing the substrate for symptoms will provide incomplete relief.
What adequate assessment looks like
Adequate assessment of new-onset anxiety in a woman in her forties or early fifties includes the hormonal context as a routine part of the clinical picture. This does not mean assuming that perimenopause explains all the symptoms. It means ensuring that the question is asked, that menstrual pattern changes are noted, that vasomotor symptoms are actively assessed, and that the timing of symptom onset is considered in relation to the perimenopausal window.
It also means holding both the psychological and the neurobiological dimensions without collapsing one into the other. A woman with a prior history of anxiety who is now in perimenopause may need both psychological support for the anxiety patterns she has carried for years, and hormonal stabilisation that reduces the neurobiological amplification of those patterns. These are not competing treatments. They are complementary ones.
What to ask if standard treatment isn't working
For women in their forties who are receiving anxiety treatment that is providing only partial relief - particularly if panic attacks are arriving without clear psychological triggers, or if there is a pattern of temperature dysregulation accompanying the anxiety - it is worth asking directly whether the hormonal context has been assessed.
The question is not whether the anxiety is "real." It is. The question is what is driving it, and whether the full picture is being treated.
Hormonal stabilisation during perimenopause - where clinically appropriate and after assessment of individual history and risk - has demonstrated efficacy for mood and anxiety symptoms during the transition. It is not the right choice for everyone. But it is a clinical option that deserves to be on the table, not discovered years later after exhaustive work in the wrong direction.
The women I see who have arrived at this understanding after several years of incomplete treatment are not angry at their previous clinicians. They are, more often, tired. Tired of a problem that was never correctly named, and quietly relieved when it finally is.